|Publication number||US7245710 B1|
|Application number||US 09/623,977|
|Publication date||Jul 17, 2007|
|Filing date||Apr 7, 1999|
|Priority date||Apr 8, 1998|
|Also published as||DE69903012D1, DE69903012T2, DE69942944D1, EP1070416A1, EP1070416B1, EP1070417A1, EP1070417B1, US6931123, WO1999053673A1, WO1999053674A1|
|Publication number||09623977, 623977, PCT/1999/1061, PCT/GB/1999/001061, PCT/GB/1999/01061, PCT/GB/99/001061, PCT/GB/99/01061, PCT/GB1999/001061, PCT/GB1999/01061, PCT/GB1999001061, PCT/GB199901061, PCT/GB99/001061, PCT/GB99/01061, PCT/GB99001061, PCT/GB9901061, US 7245710 B1, US 7245710B1, US-B1-7245710, US7245710 B1, US7245710B1|
|Inventors||Peter James Hughes|
|Original Assignee||British Telecommunications Public Limited Company|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (18), Referenced by (18), Classifications (19), Legal Events (3)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This invention relates to audio teleconferencing systems. These are systems in which three or more participants, each having a telephone connection, can participate in a multi-way discussion. The essential part of a teleconference system is called the conference “bridge”, and is where the audio signals from all the participants are combined. Conference bridges presently function by receiving audio from each of the participants, appropriately mixing the audio signals, and then distributing the mixed signal to each of the participants. All signal processing is concentrated in the bridge, and the result is monaural (that is, there is a single sound channel). This arrangement is shown in
An example is given in European Patent Specification 0291470. This discloses an arrangement in which some of the input symbols are inverted in phase before combining them in the return channel thus allowing the cancellation, for each user, of his own voice.
According to the invention, there is provided a teleconferencing system comprising a conference bridge having a multichannel connection to each of a plurality of terminal equipments, and at least one terminal equipment having means to separately process each channel to provide a plurality of outputs, each output representing one of the other terminal equipments. By adopting this multichannel approach, the conference environment can be tailored to the operating needs and circumstances of each individual by participants themselves.
Preferably the conference bridge comprises a concentrator, having means to identify the currently active input channels and to transmit only those active channels over the multichannel connection, together with control information identifying the transmitted channels. This reduces the capacity required by the multichannel connection. The control information identifying the active channels may be carried in a separate control channel, or as an overhead on the active subset of channels. In a preferred arrangement the channel representing a given terminal is excluded from the output provided to that terminal. This may be achieved by excluding that channel from the processing in the terminal equipment, but is preferably achieved by excluding it from the multichannel transmission from the bridge to that participant, thereby reducing further the capacity required by the multichannel connection.
Exemplary embodiments of the invention will now be described, by way of example, with reference to the drawings, in which:
In the conventional system illustrated in
The systems illustrated in
The transmission medium used for the uplink 3 and downlink 4,5 can be any suitable medium. ISDN (Integrated Services Data Network) technology or LAN (Local Area Network)—respectively public and private data networks—are the favoured transmission options since they provide adequate data rate and low latency—delays due to coding and transmission buffering. However, they are expensive and to date they have a low penetration in the market place. Internet Protocol (IP) techniques are becoming widely used, but currently suffer from poor latency and unreliable data rates. However, over the next few years rapid improvements are envisaged in this technology and it is likely to become the preferred telecommunication method. Such systems would be ideally suited to implementing this invention. The latest internet type modems provide 56 kbit/s downstream (links 4, 5), and up to 28.8 kbit/s upstream (link 3). They are low cost and are commonly included in personal computer retail packages. Ideally a system should be able to work with all of the above, and also with standard analogue PSTN for use as backup.
The output from the microphone 11 is encoded by an encoder 13 forming part of the terminal equipment 10, and transmitted over the uplink 3 to the exchange equipment 100. Here it is combined with the other input channels 21, 31 from the other participants, terminals into a concentrator 230 which combines the various inputs into an audio signal having a smaller number of channels 51, 52 etc. These channels are transmitted over multiple-channel digital audio links 5 to the customer equipments 10, (20, 30) where they are first decoded by respective decoders 14, 24, 34 (
The concentrator 230 selects from the input channels 11, 21, 31 those carrying useful information—typically those carrying speech—and passes only these channels over the return link 5. This reduces the amount of information to be carried. A control channel 4 carries data identifying which channels were selected. In the terminal equipment the spatialiser 15 uses data from the control channel to identify which of the original sound channels 11, 21, 31 it is receiving, and on which of the “N” channels 51, 52 in the audio link each original channel is present, and constructs a spatialised signal using that information. The spatialised signal can be tailored to the individual customer, for example the number of talkers in the spatialised system, the customer's preferences as to where in the spatialised system each participant is to seem to be located, and which channels to include.
In particular, the user may exclude the channel representing his own input 11, or may select a simultaneous translation instead of the original talker.
Transmission efficiency is achieved because only the active subset N of the total number of channels M are transmitted at any one time. The subset is chosen using a voice controlled dynamic channel allocation algorithm in the N:M concentrator 230. A possible implementation of this is shown in
A voting algorithm 234 then selects which of the inputs 11, 21, 31 have the clearest speech signals and controls a switch to direct each of the input channels 11, 21, 31 which have been selected to a respective one of the output channels 51, 52. Similar algorithms are used in Digital Circuit Multiplication Equipment (DCME) systems in international telephony circuits. Data relating the audio channels' content to the conference participants, and therefore the correspondence between the input channels 11, 21, 31 and output channels 51, 52 is transmitted over the control channel 4. Alternatively, this data can be embedded in the encoded audio data.
When there are fewer talkers identified than there are available output channels 51, 52, signal quality can be improved by using a less compressed digitisation scheme for those input channels selected, thereby using more than one output channel 51, 52 for each input channel selected. Telephone quality speech may be achieved at 8 kbits/s, allowing eight talkers to be accommodated if the system has a 64 kbit/sec capability. Should fewer talkers be detected, the 64 kbit/s capability may be used instead to provide four 16 kbit/s audio channels, capable of carrying ‘good’ quality speech, or a mixture of channels at different bit rates, to allow the coding rates to be selected according to the initial signal quality, or so that the main talker may be passed at higher quality than the other talkers.
Layered coding schemes can be used to allow graceful switching between data rates.
The N-channel de-multiplexer and speech decoder 14 used in the terminal equipment 10 is shown in
A composite signal consisting of the summation of all input signals could also be transmitted. Such a signal could be used by users having monaural receiving equipment, and may also be used by the spatialiser to generate an ambient background signal. Alternatively the spatialiser 15 may replace any of the channels 11, 21, 31 not selected by the concentrator 230, and therefore not represented in the N channel link 5, by “comfort noise”; that is low-level white noise, to avoid the aural impression of a void which would be occasioned by a complete absence of signal.
The customer equipment 10 can be implemented using a desktop PC. Readily available PC sound card technology can provide the audio interface and processing needed for the simpler spatialising schemes. The more advanced sound cards with built in DSP technology could be used for more advanced schemes. The spatialiser 15 can use any one of a number of established techniques for creating an artificial audio environment as detailed in the Hollier et al article already referred to.
Spatialisation techniques may be summarised as follows. Any of these spatialisation techniques may be used with the present invention.
The simplest technique is “panning”, where each signal is replayed with appropriate weighting via two or more loudspeakers such that it is perceived as emanating from the required direction. This is easy to implement, robust and may also be used with headphones.
“Ambisonic” systems are more complicated and employ a technique known as wavefront reconstruction to provide a realistic spatial audio percept. They can create very good spatial sound, but only for a very small listening area and are thus only appropriate for single listeners.
For headphone listening, “binaural” techniques can be used to provide very good spatialisaton. These use head-related transfer function (HRTF) filter pairs to recreate the soundfield that would have been present at the entrance to the ear canal for a sound emanating from any position in 3D space. This can give very good spatialisation and may be extended for use with loudspeakers, when it is known as “transaural”. As with ambisonic systems, the correct listening region is very small.
The output of several spatialisers may be combined as shown in
The positions of virtual sources in three dimensional space could be determined automatically, or by manual control, with the user selecting the preferred positioning for each virtual sound source. For a video conference the positioning can be set to correspond with the appropriate video picture window. The video images may be sent by other means, or may be static images retrieved from local storage by the individual user.
If the spatialised sound is relayed via loudspeakers 12, rather than headphones, it will be necessary to prevent signals from the loudspeakers 12 being picked up by the microphone 11, re-transmitted and being heard as an echo at the distant sites 20, 30 etc. A technique for achieving this will be described later, with reference to
Conventional analogue connections could be included in the conference by providing each analogue connection 43, 45 to the ‘bridge’ 200 with an encoder 42, as shown in
The invention could be applied to a conference situation in which there are several participants at each location, such as the video conference shown in
If loudspeakers are used in the embodiments of
Multi-channel echo cancellation, as shown in
The system described above with reference to
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US3825692||Sep 29, 1972||Jul 23, 1974||Stromberg Carlson Corp||Simplified conference facility for pbx|
|US4734934 *||Nov 24, 1986||Mar 29, 1988||Gte Laboratories Incorporated||Binaural teleconferencing system|
|US5020098 *||Nov 3, 1989||May 28, 1991||At&T Bell Laboratories||Telephone conferencing arrangement|
|US5307415||Oct 28, 1992||Apr 26, 1994||Fosgate James W||Surround processor with antiphase blending and panorama control circuitry|
|US5491743 *||May 24, 1994||Feb 13, 1996||International Business Machines Corporation||Virtual conference system and terminal apparatus therefor|
|US5533129||Aug 24, 1994||Jul 2, 1996||Gefvert; Herbert I.||Multi-dimensional sound reproduction system|
|US5596644 *||Oct 27, 1994||Jan 21, 1997||Aureal Semiconductor Inc.||Method and apparatus for efficient presentation of high-quality three-dimensional audio|
|US5617539 *||Jun 7, 1996||Apr 1, 1997||Vicor, Inc.||Multimedia collaboration system with separate data network and A/V network controlled by information transmitting on the data network|
|US5802180 *||Jan 17, 1997||Sep 1, 1998||Aureal Semiconductor Inc.||Method and apparatus for efficient presentation of high-quality three-dimensional audio including ambient effects|
|US5953049 *||Aug 2, 1996||Sep 14, 1999||Lucent Technologies Inc.||Adaptive audio delay control for multimedia conferencing|
|US6125115 *||Feb 12, 1998||Sep 26, 2000||Qsound Labs, Inc.||Teleconferencing method and apparatus with three-dimensional sound positioning|
|US6134223 *||Sep 18, 1996||Oct 17, 2000||Motorola, Inc.||Videophone apparatus, method and system for audio and video conferencing and telephony|
|US6327567 *||Feb 10, 1999||Dec 4, 2001||Telefonaktiebolaget L M Ericsson (Publ)||Method and system for providing spatialized audio in conference calls|
|US7012630 *||Feb 8, 1996||Mar 14, 2006||Verizon Services Corp.||Spatial sound conference system and apparatus|
|US20030081115 *||Feb 8, 1996||May 1, 2003||James E. Curry||Spatial sound conference system and apparatus|
|EP0291470A2||Apr 29, 1988||Nov 17, 1988||SGS-THOMSON MICROELECTRONICS S.r.l.||A method of attenuating the echo signal in so-called telephone fork converting circuits intended for conference, and an apparatus operating on said method|
|EP0561133A1||Feb 1, 1993||Sep 22, 1993||Nec Corporation||Multi-channel echo cancellation with adaptive filters having selectable coefficient vectors|
|EP0766446A2||Sep 25, 1996||Apr 2, 1997||Nippon Telegraph And Telephone Corporation||Method and apparatus for multi-channel acoustic echo cancellation|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7660425||May 18, 2000||Feb 9, 2010||British Telecommunications Plc||Acoustic echo cancellation|
|US7843486 *||Nov 30, 2010||Avaya Inc.||Selective muting for conference call participants|
|US7903137 *||Apr 17, 2006||Mar 8, 2011||Lifesize Communications, Inc.||Videoconferencing echo cancellers|
|US8144854 *||Mar 27, 2012||Polycom Inc.||Conference bridge which detects control information embedded in audio information to prioritize operations|
|US8175244||Dec 8, 2011||May 8, 2012||Frankel David P||Method and system for tele-conferencing with simultaneous interpretation and automatic floor control|
|US8228363||Aug 17, 2009||Jul 24, 2012||Polycom, Inc.||Method and system for conducting continuous presence conferences|
|US8494841 *||Oct 9, 2008||Jul 23, 2013||Telefonaktiebolaget Lm Ericsson (Publ)||Common scene based conference system|
|US8699716 *||Jul 7, 2004||Apr 15, 2014||Siemens Enterprise Communications Gmbh & Co. Kg||Conference device and method for multi-point communication|
|US8873669 *||Dec 20, 2004||Oct 28, 2014||Apple Inc.||Adaptable codec for selecting encoding scheme based on quality of experience|
|US9031827||Nov 30, 2012||May 12, 2015||Zip DX LLC||Multi-lingual conference bridge with cues and method of use|
|US20020103554 *||Jan 29, 2002||Aug 1, 2002||Hewlett-Packard Company||Interactive audio system|
|US20050018039 *||Jul 7, 2004||Jan 27, 2005||Gonzalo Lucioni||Conference device and method for multi-point communication|
|US20050213734 *||Mar 15, 2005||Sep 29, 2005||Polycom, Inc.||Conference bridge which detects control information embedded in audio information to prioritize operations|
|US20060239443 *||Apr 17, 2006||Oct 26, 2006||Oxford William V||Videoconferencing echo cancellers|
|US20070025538 *||Jul 11, 2005||Feb 1, 2007||Nokia Corporation||Spatialization arrangement for conference call|
|US20100194847 *||Aug 17, 2009||Aug 5, 2010||Polycom, Inc.||Method and System for Conducting Continuous Presence Conferences|
|US20110026745 *||Feb 3, 2011||Amir Said||Distributed signal processing of immersive three-dimensional sound for audio conferences|
|US20110196682 *||Oct 9, 2008||Aug 11, 2011||Telefonaktiebolaget Lm Ericsson (Publ)||Common Scene Based Conference System|
|U.S. Classification||379/202.01, 379/387.01, 379/93.21|
|International Classification||H04M3/56, H04M9/08, H04M3/36, H04M3/42, H04M3/00|
|Cooperative Classification||H04M3/002, H04M3/567, H04M3/56, H04R27/00, H04R3/02, H04M9/082|
|European Classification||H04R27/00, H04R3/02, H04M9/08C, H04M3/56, H04M3/00E|
|Sep 11, 2000||AS||Assignment|
Owner name: BRITISH TELECOMMUNICATIONS PUBLIC LIMITED COMPANY,
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:HUGHES, PETER J.;REEL/FRAME:011101/0476
Effective date: 19990413
|Jan 12, 2011||FPAY||Fee payment|
Year of fee payment: 4
|Jan 8, 2015||FPAY||Fee payment|
Year of fee payment: 8